Abstract:Sparse and asynchronous sensing and processing in natural organisms lead to ultra low-latency and energy-efficient perception. Event cameras, known as neuromorphic vision sensors, are designed to mimic these characteristics. However, fully utilizing the sparse and asynchronous event stream remains challenging. Influenced by the mature algorithms of standard cameras, most existing event-based algorithms still rely on the "group of events" processing paradigm (e.g., event frames, 3D voxels) when handling event streams. This paradigm encounters issues such as feature loss, event stacking, and high computational burden, which deviates from the intended purpose of event cameras. To address these issues, we propose a fully asynchronous neuromorphic paradigm that integrates event cameras, spiking networks, and neuromorphic processors (Intel Loihi). This paradigm can faithfully process each event asynchronously as it arrives, mimicking the spike-driven signal processing in biological brains. We compare the proposed paradigm with the existing "group of events" processing paradigm in detail on the real mobile robot dodging task. Experimental results show that our scheme exhibits better robustness than frame-based methods with different time windows and light conditions. Additionally, the energy consumption per inference of our scheme on the embedded Loihi processor is only 4.30% of that of the event spike tensor method on NVIDIA Jetson Orin NX with energy-saving mode, and 1.64% of that of the event frame method on the same neuromorphic processor. As far as we know, this is the first time that a fully asynchronous neuromorphic paradigm has been implemented for solving sequential tasks on real mobile robot.
Abstract:Event cameras have the potential to revolutionize the field of robot vision, particularly in areas like stereo disparity estimation, owing to their high temporal resolution and high dynamic range. Many studies use deep learning for event camera stereo disparity estimation. However, these methods fail to fully exploit the temporal information in the event stream to acquire clear event representations. Additionally, there is room for further reduction in pixel shifts in the feature maps before constructing the cost volume. In this paper, we propose EV-MGDispNet, a novel event-based stereo disparity estimation method. Firstly, we propose an edge-aware aggregation (EAA) module, which fuses event frames and motion confidence maps to generate a novel clear event representation. Then, we propose a motion-guided attention (MGA) module, where motion confidence maps utilize deformable transformer encoders to enhance the feature map with more accurate edges. Finally, we also add a census left-right consistency loss function to enhance the left-right consistency of stereo event representation. Through conducting experiments within challenging real-world driving scenarios, we validate that our method outperforms currently known state-of-the-art methods in terms of mean absolute error (MAE) and root mean square error (RMSE) metrics.
Abstract:Event cameras have been successfully applied to visual place recognition (VPR) tasks by using deep artificial neural networks (ANNs) in recent years. However, previously proposed deep ANN architectures are often unable to harness the abundant temporal information presented in event streams. In contrast, deep spiking networks exhibit more intricate spatiotemporal dynamics and are inherently well-suited to process sparse asynchronous event streams. Unfortunately, directly inputting temporal-dense event volumes into the spiking network introduces excessive time steps, resulting in prohibitively high training costs for large-scale VPR tasks. To address the aforementioned issues, we propose a novel deep spiking network architecture called Spike-EVPR for event-based VPR tasks. First, we introduce two novel event representations tailored for SNN to fully exploit the spatio-temporal information from the event streams, and reduce the video memory occupation during training as much as possible. Then, to exploit the full potential of these two representations, we construct a Bifurcated Spike Residual Encoder (BSR-Encoder) with powerful representational capabilities to better extract the high-level features from the two event representations. Next, we introduce a Shared & Specific Descriptor Extractor (SSD-Extractor). This module is designed to extract features shared between the two representations and features specific to each. Finally, we propose a Cross-Descriptor Aggregation Module (CDA-Module) that fuses the above three features to generate a refined, robust global descriptor of the scene. Our experimental results indicate the superior performance of our Spike-EVPR compared to several existing EVPR pipelines on Brisbane-Event-VPR and DDD20 datasets, with the average Recall@1 increased by 7.61% on Brisbane and 13.20% on DDD20.
Abstract:Traditional visual place recognition (VPR), usually using standard cameras, is easy to fail due to glare or high-speed motion. By contrast, event cameras have the advantages of low latency, high temporal resolution, and high dynamic range, which can deal with the above issues. Nevertheless, event cameras are prone to failure in weakly textured or motionless scenes, while standard cameras can still provide appearance information in this case. Thus, exploiting the complementarity of standard cameras and event cameras can effectively improve the performance of VPR algorithms. In the paper, we propose FE-Fusion-VPR, an attention-based multi-scale network architecture for VPR by fusing frames and events. First, the intensity frame and event volume are fed into the two-stream feature extraction network for shallow feature fusion. Next, the three-scale features are obtained through the multi-scale fusion network and aggregated into three sub-descriptors using the VLAD layer. Finally, the weight of each sub-descriptor is learned through the descriptor re-weighting network to obtain the final refined descriptor. Experimental results show that on the Brisbane-Event-VPR and DDD20 datasets, the Recall@1 of our FE-Fusion-VPR is 29.26% and 33.59% higher than Event-VPR and Ensemble-EventVPR, and is 7.00% and 14.15% higher than MultiRes-NetVLAD and NetVLAD. To our knowledge, this is the first end-to-end network that goes beyond the existing event-based and frame-based SOTA methods to fuse frame and events directly for VPR.
Abstract:Traditional visual navigation methods of micro aerial vehicle (MAV) usually calculate a passable path that satisfies the constraints depending on a prior map. However, these methods have issues such as high demand for computing resources and poor robustness in face of unfamiliar environments. Aiming to solve the above problems, we propose a neuromorphic reinforcement learning method (Neuro-Planner) that combines spiking neural network (SNN) and deep reinforcement learning (DRL) to realize MAV 3D visual navigation with depth camera. Specifically, we design spiking actor network based on two-state LIF (TS-LIF) neurons and its encoding-decoding schemes for efficient inference. Then our improved hybrid deep deterministic policy gradient (HDDPG) and TS-LIF-based spatio-temporal back propagation (STBP) algorithms are used as the training framework for actor-critic network architecture. To verify the effectiveness of the proposed Neuro-Planner, we carry out detailed comparison experiments with various SNN training algorithm (STBP, BPTT and SLAYER) in the software-in-the-loop (SITL) simulation framework. The navigation success rate of our HDDPG-STBP is 4.3\% and 5.3\% higher than that of the original DDPG in the two evaluation environments. To the best of our knowledge, this is the first work combining neuromorphic computing and deep reinforcement learning for MAV 3D visual navigation task.
Abstract:We develop a deep convolutional neural network (DCNN) based framework for model-free prediction of the occurrence of extreme events both in time ("when") and in space ("where") in nonlinear physical systems of spatial dimension two. The measurements or data are a set of two-dimensional snapshots or images. For a desired time horizon of prediction, a proper labeling scheme can be designated to enable successful training of the DCNN and subsequent prediction of extreme events in time. Given that an extreme event has been predicted to occur within the time horizon, a space-based labeling scheme can be applied to predict, within certain resolution, the location at which the event will occur. We use synthetic data from the 2D complex Ginzburg-Landau equation and empirical wind speed data of the North Atlantic ocean to demonstrate and validate our machine-learning based prediction framework. The trade-offs among the prediction horizon, spatial resolution, and accuracy are illustrated, and the detrimental effect of spatially biased occurrence of extreme event on prediction accuracy is discussed. The deep learning framework is viable for predicting extreme events in the real world.
Abstract:Reservoir computing systems, a class of recurrent neural networks, have recently been exploited for model-free, data-based prediction of the state evolution of a variety of chaotic dynamical systems. The prediction horizon demonstrated has been about half dozen Lyapunov time. Is it possible to significantly extend the prediction time beyond what has been achieved so far? We articulate a scheme incorporating time-dependent but sparse data inputs into reservoir computing and demonstrate that such rare "updates" of the actual state practically enable an arbitrarily long prediction horizon for a variety of chaotic systems. A physical understanding based on the theory of temporal synchronization is developed.
Abstract:A common difficulty in applications of machine learning is the lack of any general principle for guiding the choices of key parameters of the underlying neural network. Focusing on a class of recurrent neural networks - reservoir computing systems that have recently been exploited for model-free prediction of nonlinear dynamical systems, we uncover a surprising phenomenon: the emergence of an interval in the spectral radius of the neural network in which the prediction error is minimized. In a three-dimensional representation of the error versus time and spectral radius, the interval corresponds to the bottom region of a "valley." Such a valley arises for a variety of spatiotemporal dynamical systems described by nonlinear partial differential equations, regardless of the structure and the edge-weight distribution of the underlying reservoir network. We also find that, while the particular location and size of the valley would depend on the details of the target system to be predicted, the interval tends to be larger for undirected than for directed networks. The valley phenomenon can be beneficial to the design of optimal reservoir computing, representing a small step forward in understanding these machine-learning systems.